[dev.icinga.com #9437] Notifications are not working properly when enabled in master- and child-zone

[icinga-migration]

This issue has been migrated from Redmine: https://dev.icinga.com/issues/9437

Created by mwaldmueller on 2015-06-17 10:53:07 +00:00

Assignee: (none)
Status: Feedback
Target Version: (none)
Last Update: 2016-11-09 14:55:34 +00:00 (in Redmine)

Icinga Version: 2.3.4
Backport?: Not yet backported
Include in Changelog: 1

---

My setup:

• "master"-zone is parent zone with one node
• "checker"-zone is child zone with one node
• notification feature is enabled on all nodes in both zones

Problem: Inital notification is send from both zones
Problem: Re-notifications are either send from master- or checker-zone

---

Relations:

• relates #11562

[icinga-migration]

Updated by mfriedrich on 2015-07-24 18:03:07 +00:00

• Target Version set to Backlog

[icinga-migration]

Updated by mj84 on 2016-04-06 11:13:02 +00:00

Are there any news on this one? Any chance this might be included with #11311?

[icinga-migration]

Updated by mfriedrich on 2016-04-06 15:34:56 +00:00

• Parent Id set to 11311

No, sorry. Though I missed this one when collecting it into the parent task, thanks for the reminder.

[icinga-migration]

Updated by mfriedrich on 2016-08-02 13:29:05 +00:00

• Relates set to 11562

[icinga-migration]

Updated by mfriedrich on 2016-08-02 13:29:41 +00:00

You could test the snapshot packages which involves changes for the cluster protocol now syncing the notification timestamps.

[icinga-migration]

Updated by mj84 on 2016-08-02 16:49:04 +00:00

Unfortunately I do not have any master/child setup in place which allows the installation of snapshot packages, so I'll provide an update on this issue once 2.5.0 is out.
Does that work for you as well?

[icinga-migration]

Updated by mfriedrich on 2016-08-16 12:51:51 +00:00

• Status changed from New to Feedback
• Target Version deleted Backlog

Sure. Please let us know then.

[icinga-migration]

Updated by mj84 on 2016-09-01 12:27:30 +00:00

Hi,

as far as I can observe, the behaviour hasn't changed in 2.5.4.
I've just re-enabled the notification feature on my master, activated my dummy host object (which is always down), and received the following alerts:

14:13:31 From satellite: PROBLEM - NotificationTest - ping4 is CRITICAL (Service notification)
14:13:31 From master: PROBLEM - NotificationTest - ping4 is CRITICAL (Service notification)
14:14:49 From satellite: PROBLEM - NotificationTest is DOWN (Host notification)
14:14:49 From master: PROBLEM - NotificationTest is DOWN (Host notification)
14:15:19 From satellite: PROBLEM - NotificationTest is DOWN (Host notification)
14:15:21 From master: PROBLEM - NotificationTest is DOWN (Host notification)
14:15:51 From master: PROBLEM - NotificationTest is DOWN (Host notification)
14:15:54 From satellite: PROBLEM - NotificationTest is DOWN (Host notification)
14:16:21 From master: PROBLEM - NotificationTest is DOWN (Host notification)
14:16:29 From satellite: PROBLEM - NotificationTest is DOWN (Host notification)
14:16:51 From master: PROBLEM - NotificationTest is DOWN (Host notification)
14:17:04 From satellite: PROBLEM - NotificationTest is DOWN (Host notification)
14:17:21 From master: PROBLEM - NotificationTest is DOWN (Host notification)
14:17:39 From satellite: PROBLEM - NotificationTest is DOWN (Host notification)

For some reason, the satellite does not send the notifications in an exact 30 seconds interval (which is configured), but the master does.
Please note, that the timestamps are taken from the notification mail body, and both system clocks are in sync.

Regards,
Markus

[icinga-migration]

Updated by mfriedrich on 2016-11-09 14:55:34 +00:00

• Parent Id deleted 11311

[Al2Klimov]

@mwaldmueller Does this issue still occur in v2.10.2?

[dnsmichi]

Markus is on vacation, there's no feedback any time soon. I'll close this and if it is still the case, please create a new issue.